|
@@ -0,0 +1,28 @@
|
|
|
|
|
+import newspaper
|
|
|
|
|
+from newspaper import Article,Config
|
|
|
|
|
+import sys
|
|
|
|
|
+import requests
|
|
|
|
|
+
|
|
|
|
|
+url = sys.argv[1]
|
|
|
|
|
+
|
|
|
|
|
+config = Config()
|
|
|
|
|
+config.language='en'
|
|
|
|
|
+config.memoize_articles=False
|
|
|
|
|
+config.keep_article_html=True
|
|
|
|
|
+config.request_timeout=10
|
|
|
|
|
+config.fetch_images=False
|
|
|
|
|
+config.browser_user_agent="Mozilla/5.0 (X11; Linux x86_64; rv:47.0) Gecko/20100101 Firefox/47.0"
|
|
|
|
|
+config.verbose=True
|
|
|
|
|
+
|
|
|
|
|
+
|
|
|
|
|
+#r = requests.get(url)
|
|
|
|
|
+#print(r.status_code)
|
|
|
|
|
+#print(r.text)
|
|
|
|
|
+a = Article(url=url,config=config)
|
|
|
|
|
+print(url)
|
|
|
|
|
+a.download()
|
|
|
|
|
+print("dl")
|
|
|
|
|
+a.parse()
|
|
|
|
|
+print("parsed")
|
|
|
|
|
+print(a.text)
|
|
|
|
|
+#
|